Raw solexa data processing

[MiniRider]

Hi there,

I am a beginner in sequencing field, I have several silly questions hopefully someone can give me some instruction.

I just got the our sequencing data from Solexa 1G analyzer which gave us s_*_000*_seq.txt and s_*_000*_prb.txt

First question:
Why the length of each read is 152bps (76x2) instead of 76bps (38x2)?

Second question:
What is the first process I should do? I already convert raw text file into .fasta format, should I split the 152bp read into 2 short reads with 76 bps or even shorter?

Third question:
For data input for Maq, we need to convert reads from fastq to bfg. But I only have s_*_000*_seq.txt and s_*_000*_prb.txt in hand, which software can I use for this step conversion or I can do this by Matlab?

Thanks your time helping me answer these questions.

[ECO]

Hey MiniRider,

For your 3rd question, see this page, which contains the following instructions:

IMPORTANT NOTE: The raw reads format used by Solexa (those `s_?_sequence.txt' from the Solexa pipeline) are different from mapass' FASTQ format in that the qualties are scaled differently. To use maq, you need to first convert the format with:

• maq sol2sanger s_1_sequence.txt s_1_sequence.fastq

where s_1_sequence.txt is the Solexa read sequence file. Missing this step will lead to unreliable SNP calling.

First you'll have to use the above advice to convert your solexa fastq to a sanger-scaled fastq, then you will simply use "maq fastq2bfq" to make the bfq files. If you run into computational issues, you can split the reads at this step using the -n option.

Hope that helps.

[kevinlu]

Hi there all,

I'm also a beginner, and quite inexperienced at that. What does it take to convert the raw eland output from the Solexa machine into a .bed format suitable for viewing purposes on something like the UCSC Genome Browser?

Thanks,
Kevin